DESCRIPTION: The mechanisms by which homologous chromosomes pair with each other remains mysterious; evidence can be mustered to support the existence or non-existence of discrete sites of pairing initiation. Historically, most interest in chromosome pairing has focused on the first meiotic division, where pairing is essential for recombination and proper chromosome segregation. Dr. Golic's lab has recently obtained a body of evidence supporting the idea that homologous chromosome pairing during mitosis serves as a close model for meiotic chromosome pairing. Most convincingly, he has in collaboration with Dr. Sergio Pimpinelli found genetic and cytological data that mitotic recombination during G2 causes homologous chromosomes to associate as a bivalent as they attach to the mitotic spindle. The basic thrust of this revised proposal is to use FLP-mediated DNA mobilization - a technique pioneered in the Golic lab - as a test system to explore the nature of homologous chromosome pairing in mitotic cells. In this approach, DNA sequences flanked by FRT sites are excised from one chromosomal location (the donor) by the action of FLP recombinase. The resulting FRT-bearing circles can reintegrate into a second chromosomal location (the target or recipient) that also contains an FRT site. In the first specific aim, Dr. Golic asks whether the original location of the donor in the genome influences the frequency of targeting. If the donor and recipient sites are far apart, will this reduce the frequency of mobilization? If this is not the case, it will imply that excised circles can diffuse freely through the nucleus. To answer this question, Dr. Golic will conduct the mobilization experiment using a variety of donor and recipient sites variably positioned on the same or different chromosomes. These experiments will utilize a sensitive somatic assay in which integration at a target site reconstructs a white+ gene, resulting in red clones in the background of a white-colored eye. The second specific aim is designed to extend previous observations that the larger the stretch of homology between the donor and target sites (outside of the FRTs they share), the more efficient the mobilization. This presumably results because homologous pairing helps bring the donor circle and target into proximity. Dr. Golic has devised a competition experiment in which cells will have two target sites with different amounts of homology with the donor. DNA from the non-preferred target site will then be added to the donor, to see if this alters target site bias. Control experiments will be performed to insure that any alteration in target site bias is due to the specific DNA sequences from the target site, rather than the length of the DNA segment employed. In the third specific aim, Dr. Golic will test whether certain candidate DNA sequences thought to be involved in pairing indeed play that role in the mobilization assay. These candidates include the PRE sequences recognized by Polycomb Group proteins, polymers of the Zeste protein binding site, repeats of the AG dinucleotide recognized by the GAGA protein (the product of the Trithorax-like gene), and other repetitive sequences. Candidate sites will be added to both donors and recipients to check for heightened rates of mobilization. Some of these experiments will be conducted as competition assays to see if candidate-containing donors will preferentially select targets with or without the candidate pairing sequence. The fourth specific aim will aim to adapt for Drosophila the "cassette exchange" method of FLP-mediated DNA integration developed by Schlake and Bode. This method employs two different FRT sites that cannot recombine with each other, so that integration of a sequences from a circular DNA molecule can occur only when two recombination events take place between FRT1 and FRT2 elements on both the donor and recipient. This technique should improve the efficiency of placing DNA at the target site because (in contrast to the current method), the integrated DNA cannot be lost subsequently by re-excision. The fifth specific aim details a clever experimental strategy which will exploit the findings of the four previous specific aims to identify DNA sequences that promote pairing with chromosomal target sites. The basis of this strategy is the injection of a library of random chromosomal fragments, cloned into an FRT cassette exchange vector, into embryos with an appropriate target construct containing a different antibiotic resistance gene. The FLP recombinase will be supplied in the form of a synthetic mRNA coinjected with the plasmid library. Plasmids containing sequences promoting plasmid/chromosome pairing should be favored for integration; these sequences should be recovered because the cassette exchange will switch the antibiotic resistance gene on the plasmid. Several rounds of this procedure should allow a high degree of enrichment of favored pairing sequences. The genomic location of DNA sequences recovered by this strategy will be determined to see if they contain a prominent pairing site or contain DNA sequences from the vicinity of the target element. The sixth and final specific aim will look at two pairing-dependent phenomena. The first of these is a variegated eye phenotype called pugilist-D, which is caused by a gene in which coding sequences for a biosynthetic enzyme are fused in frame to 1 kb of repeats of the GAGA factor binding site. Dr. Golic wishes to determine whether this pug-D phenotype results from the pairing of the fused gene back to centric heterochromatin containing the same repeated sequence. This will be done by using FLP-mediated DNA mobilization to place a pug-D transgene at different distances from centric heterochromatin. The second phenomenon explored in the last specific aim concerns the ability of a PRE sequence to silence an adjacent reporter gene in a pairing-dependent fashion. The question here is whether the silencing apparatus is reset every mitotic cycle. To this end, a construct will be made in which the PRE can be removed from its location adjacent to a white+ reporter by inducing FLP recombinase. If PRE function is indeed reset during mitosis, then the anterior region of the eye in front of the mitotic wave should show the normal variegation of the insertion (because this area is reset after the mitotic wave has passed over these cells), while the posterior region of the eye behind the mitotic wave should show strong reporter activity because silencing cannot be maintained in the absence of the PRE.